mattpocock-skills
claudeMatt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
A delivery funnel for one repository task, in three lanes by what the change can break: spec, tests before the code on risky paths, the repo's own gates, a review whose findings are verified before anything is fixed, draft pull request. Configured per repo, so the same funnel runs on any stack.
Plugin task-funnel from marketplace task-funnel in gabrielbsandim/task-funnel. Commands read from the repo's manifest — today.
This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.
backlogGrooms an open backlog before the next task is picked: finds issues that are the same work, adopts orphans into the epic they belong to, and proposes what can ship in one pull request. Read-only until the user approves each move. Use when the board feels like it only grows, before planning a sprint
investigateInvestigation flow. Takes a question about the system (why does X happen, where does Y live, is Z feasible, what broke) and returns a verified written answer with evidence. Read-only end to end, no branch, no code change, no pull request. Use when asked to investigate, diagnose, explain behaviour or
noteWrites one durable fact into a markdown knowledge vault, with the frontmatter and the provenance that keep it checkable, then reindexes so search can find it. Use when something learned in a session has to outlive it, or on /note <the fact>, and not for what the repo already records.
reviewReviews a diff, branch, commit or pull request this session did not write and returns only the Critical and Important findings that survive an attempt to refute them, each with the input and the wrong output. Use when asked, in any language, to review, check or look over a diff, a branch, a commit o
shipTakes work that is already implemented and committed on a branch and ships it: runs the repo's ship gates off receipts, pushes, opens a draft pull request, moves the board to in review, and removes the worktree. Use when the code exists and the task is to get it reviewable, or on /ship, and not when
taskDelivery funnel for one repository task: a feature, bug fix, refactor or chore, from description or issue number to a pushed branch and a draft pull request with the board moved. Use whenever the user asks, in any language, to implement, fix, build, change or refactor something in this repository, i
/task-funnel:backlogGroom the open backlog: what is the same work, what is an orphan, and what can ship in one pull request
/task-funnel:setupOne-time setup for task-funnel in this environment, then a per-repo config check
funnel-reviewerStage 4 of the task funnel. Reviews one already-written diff slice against a numbered multi-part contract, one part per lens, and returns only findings that would make the change untrustworthy, graded Critical or Important, CONFIRMED or PLAUSIBLE, and NEW or PRE-EXISTING. Reads the diff file it is g
funnel-screen-lensStage 4 of the task funnel, browser half. Opens the screens a change touched in a real browser and answers two numbered parts from one visit: whether the feature works, and whether the interface can be used. Returns only findings graded Critical or Important, each verified before it costs a fix. Rea
funnel-test-writerStage 2 of the task funnel, risk lane only. Writes the failing tests for a change before the change exists, from the spec's acceptance criteria and interface, proves they fail for the reason the criterion names, and commits them. The main session then writes the code that makes them pass. Invoked by
funnel-triageStage 1 of the task funnel, risk lane only. Read-only. Checks whether the premise of a risky task is true before anything is built on it, then writes the spec the tests and the code will be held to. Returns a structured verdict, the premise check with its evidence, file scope, acceptance criteria an
funnel-verifierStage 4 of the task funnel, second half. Takes the Critical and Important findings one review round returned and tries to refute each one against the code as committed, so that only a finding that reproduces sends the change back for a fix. Returns REAL or REFUTED per finding, with the evidence. Inv
python3 "${CLAUDE_PLUGIN_ROOT}/hooks/guards.py"
python3 "${CLAUDE_PLUGIN_ROOT}/hooks/guards.py"
python3 "${CLAUDE_PLUGIN_ROOT}/hooks/gate-receipt.py"
python3 "${CLAUDE_PLUGIN_ROOT}/hooks/wip.py"
Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
15 skills · hooks
Use Alibaba Open Code Review from local Codex.
2 skills
Backend API design, GraphQL architecture, workflow orchestration with Temporal, and test-driven backend development
9 skills · 1 command · 8 agents
AI-supervised issue tracker for coding workflows
1 skill · 1 agent · hooks
MiniMax AI skills library for frontend, fullstack, Android, iOS, shader, GIF sticker, document, presentation, spreadsheet, and multimodal media workflows
18 skills
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.